*
*The first analysis uses the dataset and looks for the effect of the variables upon the likelihood of holding a hearing on an explicitly racial issue.
logit hearing AA_chair AA_num Latino_chair Latino_num CmteMemNum redist post_reform dem_house aa latinx ag_cmte wam_cmte smbus_cmte jud_cmte ed_cmte banking_cmte budget_cmte comm_cmte 
*
*
*
*This analysis replaces the committee dummies with Deering and Smith's (1984) committee type dummies.
logit hearing AA_chair AA_num Latino_chair Latino_num CmteMemNum redist post_reform dem_house aa latinx service_dummy constituency_dummy policy_dummy prestige_dummy
*This uses Clarify to simulate moving the number of Black committee members from their minimum to their maximum on committee types.
*Black committee member simulations, by committee type
*Policy Type-Committees
estsimp logit hearing AA_chair Latino_chair AA_num Latino_num CmteMemNum redist post_reform dem_house aa latinx service_dummy constituency_dummy policy_dummy prestige_dummy, genname(polAA)
help setx 
setx AA_chair 0 Latino_chair 0 AA_num 0 Latino_num 0 CmteMemNum mean redist 1 post_reform 1 dem_house 1 aa 47 latinx 34 policy_dummy 1 service_dummy 0 constituency_dummy 0 prestige_dummy 0
simqi
setx AA_chair 0 Latino_chair 0 AA_num 13 Latino_num 0 CmteMemNum mean redist 1 post_reform 1 dem_house 1 aa 47 latinx 34 policy_dummy 1 service_dummy 0 constituency_dummy 0 prestige_dummy 0
simqi
*Service Type-Committees
estsimp logit hearing AA_chair Latino_chair AA_num Latino_num CmteMemNum redist post_reform dem_house aa latinx service_dummy constituency_dummy policy_dummy prestige_dummy, genname(servAA)
help setx 
setx AA_chair 0 Latino_chair 0 AA_num 0 Latino_num 0 CmteMemNum mean redist 1 dem_house 1 aa 47 latinx 34 service_dummy 1 policy_dummy 0 constituency_dummy 0 prestige_dummy 0
simqi
setx AA_chair 0 Latino_chair 0 AA_num 7 Latino_num 0 CmteMemNum mean redist 1 dem_house 1 aa 47 latinx 34 service_dummy 1 policy_dummy 0 constituency_dummy 0 prestige_dummy 0
simqi
*Constituency Type-Committees
estsimp logit hearing AA_chair Latino_chair AA_num Latino_num CmteMemNum redist post_reform dem_house aa latinx service_dummy constituency_dummy policy_dummy prestige_dummy, genname(conAA)
help setx 
setx AA_chair 0 Latino_chair 0 AA_num 0 Latino_num 0 CmteMemNum mean redist 1 dem_house 1 aa 47 latinx 34 constituency_dummy 1 service_dummy 0 prestige_dummy 0 policy_dummy 0
simqi
setx AA_chair 0 Latino_chair 0 AA_num 13 Latino_num 0 CmteMemNum mean redist 1 dem_house 1 aa 47 latinx 34 constituency_dummy 1 service_dummy 0 prestige_dummy 0 policy_dummy 0
simqi
*Prestige Type-Committees
estsimp logit hearing AA_chair Latino_chair AA_num Latino_num CmteMemNum redist post_reform dem_house aa latinx service_dummy constituency_dummy policy_dummy prestige_dummy, genname(presAA)
help setx 
setx AA_chair 0 Latino_chair 0 AA_num 0 Latino_num 0 CmteMemNum mean redist 1 dem_house 1 aa 47 latinx 34 prestige_dummy 1 constituency_dummy 0 service_dummy 0 policy_dummy 0
simqi
setx AA_chair 0 Latino_chair 0 AA_num 7 Latino_num 0 CmteMemNum mean redist 1 dem_house 1 aa 47 latinx 34 prestige_dummy 1 constituency_dummy 0 service_dummy 0 policy_dummy 0
simqi
*Proportions test using results from simulations, to see if the differences when raising the number of Black committee members from their minimums to their
*maximums are significant
*policy
prtesti 1000 0.72 1000 0.97
*service
prtesti 1000 0.2 1000 0.51
*constituency
prtesti 1000 0.66 1000 0.96
*prestige
prtesti 1000 0.61 1000 0.87
